Some common methods for estimating rotating equipment component service life

How long will your machines be reliable in service is a question whose answer every equipment owner wants to know. Reliability engineering has a number of ways to arrive at an estimate for the length of a part’s service life before failure

 


Hello Mike,

I am interested in determining mechanical design life for large rotating equipment used in manufacture of fibre optic cable. In most cases the machines consist of weldments spaced between turning rings supported by a roller undercarriage.

Do you have technical guides or services to perform these calculations? In general, I am more concerned with fatigue of the weldments than general power transmission components.

My second area of criticality would be large gearboxes and transmissions (symptoms of fatigue vs gear wear).

Please advise if these need areas are within your scope of services / information.

 


 

Dear Ric,

Estimating design life of machinery is always a best guess answer. The length of rotating equipment working life totally depends on what material its components are made of, and what stresses they suffered during their lifetimes, starting from their fabrication. As parts are made the processes used in their manufacture induce stresses and microcracks. Flame cutting, grinding, rough machining of surfaces, pressed bends and shapes, along with all other manufacturing techniques that involve plastic deformation of the material, damage microstructures. The damage becomes a defect that will start a future failure under the wrong circumstances. You must also consider if the part’s microstructure will be chemically or physically destroyed during service so that its material of construction will disappear, such as from corrosion, abrasion, attack by the product, etc.

It’s more meaningful to ask how long you need the machinery to remain in operation under the conditions and service duty you require it to operate. Then a yes or no answer for that duration can be determined. The answer will be based on what has happened to other similar equipment in similar situations elsewhere in the world. You would look at ‘real life’ use of the equipment and apply those results to your situation to arrive at some sort of believable guesstimate to work with. Some industries, such as oil and gas, and electronics, have developed data bases of equipment failures to permit estimations of operating life.

Another good resource to ask how long parts might last in a particular service is the original equipment manufacturer. Particularly for gearboxes. If the OEM has been in business for several decades, and they have stayed in touch with their clients, they will have a recorded history of problems suffered by their designs. From the failure history data they can sensibly guesstimate their machinery’s’ life in your operation.

There has been research done on the fatigue life of weldments. Weibull Analysis calculations were invented as a result of weldment reliability research. The equations require failure data in order to determine the reliability of other identical items in identical duty conditions. This means you have to fail a sample part several times to get its lifetime to failure. If you change the working conditions, or material, or design shape you have to do new trials.

Your most certain option is to work with a metallurgical laboratory using samples of your welded parts. They can run accelerated fatigue life tests and determine a range of working life for when the parts are in service in your operation.

Another way to get a life estimate of components is to use the published fatigue curves for their materials of construction. You would need to calculate the forces in the part at various locations to find the points of highest stress a component would suffer. The effect of stress raising shape changes would need to be factored into the analysis. From the information on part geometry and operating loads you calcualte a stress to estimate off the material’s fatigue curve a number of cycles to failure.

You would need fully dimensioned drawings and the full range of service loads the part will experience to calculate stresses throughout the part. A finite element analysis (FEA) computer model of the component would be drawn and the stresses in the material from the working loads would be calculated. The FEA model would show the locations of maximum stress and the values. The highest stress would be the entry point on a fatigue curve from which you would get a value for cycles to failure at that stress level.

The cycles to failure at maximum operating stress identified from the fatigue curve would indicate the part’s operating life. This method is theoretic-based using published engineering data and calculated loads and its results are still a guesstimate. It is a simplistic analysis because the stress applied on the model may not be the stress the part suffers in real life. But it does give you a means to work from the design drawings to arrive at a not unrealistic estimate for a part’s operating service life.

Don’t be seduced by such possibilities as accelerated-life laboratory testing and FEA computer modelling into thinking your lifetime analysis of parts will mimic the real world situations your machines will suffer. In the real world your machines will be overload by bad operating practices. They will be installed with soft foot distortion. They undergo rapid changes in operating conditions, like temperature, throughputs and pressure. Parts will be mounted by hammering and forcing them into place. Lubrication oil will get contaminated with wear particles. There will be product ingress, moisture ingress, and wrong lubricants inserted into the equipment. The manufacturer will make fabrication and machining errors. Parts will be installed with the wrong clearances, either too loose or too tight. Plus a hundred other ways to break your machines will happen to them!

Your machines will most likely never live their full theoretic service life because their parts will be destroyed by the combined sudden changes, mistakes and errors done to them as they are made and used.

To get the full service life from your machines and equipment that is potentially possible, you need an enterprise asset management system that controls their parts’ operating conditions, service loads, and microstructure health, so your company’s bad practices and its peoples’ ignorance don’t destroy your equipment.

The only solution I know that builds a life cycle asset management system that gets the maximum reliability and failure-free lifetime from your equipment is the Plant Wellness Way asset management methodology.

I hope that the above is useful information for you.

 

All the best to you,

Mike Sondalini
Managing Director
Lifetime Reliability Solutions HQ